A conventional cluster storage system includes a computing device and at least one storage device connected to the computing device. The computing device is installed with Linux Operation System (OS) for controlling operations of the computing device and the storage device.
The storage device, for example, adopts an architecture of just a bunch of disks (JBOD), and includes a plurality of expanders each connected to a plurality of hard drives, and a plurality of slots for installation of the respective hard drives. In Linux OS, each of the hard drives corresponds to at least one logical device name. The storage device further includes a plurality of warning lights each corresponding to a respective one of the slots. When an abnormal hard drive among the hard drives is detected by firmware of the storage device, one of the warning lights corresponding to the slot installed with the abnormal hard drive is triggered to emit light, so as to notify a system administrator about which one of the slots the abnormal hard drive is installed. In this way, the system administrator is allowed to retrieve the abnormal hard drive from the storage device.
However, it is insufficient to rely on detecting the abnormal hard drive merely by the firmware itself in the storage device. Since the firmware of the storage device adopts relatively strict standard for determination of an abnormal hard drive, only the hard drive that is not able to operate would be determined as an abnormal hard drive by the firmware of the storage device.
Aside from detecting an abnormal hard drive by virtue of the firmware of the storage device, currently, the Linux OS installed in the computing device or a third-party computer program executed in the computing device may be utilized to determine whether the hard drives of the storage device have any abnormal condition other than abnormal conditions that are able to be detected by the firmware of the storage device. For example, the third-party computer program is executed to perform read/write test upon the hard drives, and to determine the hard drive which is unable to pass the read/write test as an abnormal hard drive. However, conventionally, when the aforementioned Linux OS or the third-party computer program detects an abnormal hard drive, only the logical device name corresponding to the abnormal hard drive thus detected is outputted. Since the system administrator is not able to recognize which physical slot the abnormal hard drive is installed in, the abnormal hard drive cannot be retrieved easily for maintenance or replacement.